Search bioRxiv⌕ Search

Biology subjects

Loyfer, N.

Publications and source records attributed to Loyfer, N..

13 recordsLinked to original sources

Ultrasensitive detection of neuronal DNA using genome-wide non-CG methylation patterns

Methylation-based liquid biopsies can identify the tissue origins of circulating cell-free DNA (cfDNA). A major constraint on the sensitivity of these assays is the limited number of cell-type-specific CpG markers. Here we exploit the high abundance of non-CpG (CpH) methylation in neurons to enable ultrasensitive detection of neuronal DNA. We show that neurons carry 100-fold higher CpH methylation than other tissues, in which apparent CpH methylation largely reflects experimental artifacts. Using a computational criterion to identify cfDNA molecules carrying clusters of methylated CpA sites, we detect neuronal DNA at proportions as low as 20 parts per million. Analysis of 231 samples from 163 individuals revealed minimal neuronal cfDNA in the plasma of healthy subjects; elevated levels in patients with measles or acute neurological conditions; strikingly high levels in brain-dead donors, indicating extensive ongoing neuronal death; and the highest levels in the cerebrospinal fluid of patients in a neurological intensive care unit. These findings establish genome-wide, fragment-level analysis of non-CpG methylation as an ultrasensitive approach for blood-based monitoring of brain injury and disease.

genomics↗

Systematic errors in enzymatic conversion limit cell-free DNA methylation specificity

Enzymatic methylation sequencing (EM-seq) converts unmethylated cytosines to uracils while preserving DNA integrity, making it attractive for liquid biopsies. Here we report a reproducible fragment-level over-conversion error in EM-seq, which is not observed in bisulfite-based conversion or in Oxford Nanopore sequencing. While bisulfite and nanopore errors occur at sporadic CpGs, EM-seq generates molecules that appear fully unmethylated, introducing a false-positive background signal that severely limits deconvolution specificity in cfDNA analysis.

genomics↗

The genetic basis for DNA methylation variation across tissues and development

The mechanisms by which genetic variation shapes the epigenome across cell types and developmental stages have remained elusive. Here, we define a unifying developmental framework for DNA methylation programming, grounded in genome-wide methylation and genetic variation data from both mouse and human. In mice, we identify thousands of differentially methylated regions (DMRs) linked to sequence polymorphisms that disrupt transcription factor binding. These DMRs are programmed either during implantation or later in organogenesis, revealing two distinct periods of epigenetic regulation. Extending this logic to humans, we analyze our atlas of over 200 WGBS samples from 39 purified cell types and map 33,574 regions where common SNPs control allele-specific methylation. These include both early-established and cell-type-specific loci, many of which colocalize with eQTLs, enhancers, silencers, and disease-associated variants. Our results uncover a widespread mechanism by which genetic variation influences the regulatory landscape, linking sequence, methylation, and transcription across tissues. This cross-species atlas of sequence-dependent methylation not only clarifies the logic and timing of epigenetic programming, but also provides a foundational resource for deciphering non-coding variants in development, complex disease, and regenerative medicine.

genomics↗

Human whole epigenome modelling for clinical applications with Pleiades

Gene regulation in humans extends beyond the four-letter genetic code. DNA methylation, in particular, functions as a critical epigenetic regulator, dynamically programming cellular identity, adapting gene expression in response to environmental cues, and underpinning the onset and progression of numerous diseases. Here we present Pleiades, a series of whole-genome epigenetic foundation models spanning three sizes: 90M, 600M, and 7B parameters. Pleiades is trained upon an extensive proprietary corpus of human methylation and genomic data, totalling 1.9T tokens. We introduce alignment embeddings and stacked hierarchical attention techniques to provide precise epigenetic modelling without the need for extended context lengths. Collectively, these advances enable Pleiades to perform a diverse range of down-stream biological and clinical tasks, including genomic regulatory prediction, realistic generation of cell-free DNA fragments and fragment-level cell-type-of-origin classification, within a unified and scalable computational framework. We specifically apply Pleiades to the early detection of clinical Alzheimers disease and Parkinsons disease from plasma cell-free DNA, achieving high-accuracy detection (AUROC 0.89 for AD and 0.84 for PD) using a minimally invasive blood test. Combined with plasma pTau-217, Pleiades reaches an AUROC of 0.97 for AD, underscoring the promise of multimodal epigenomic and proteomic approaches. Using mechanistic interpretability, we ground Pleiades latent features in interpretable biological signals, relating its Alzheimers predictions to fragmentomic and epigenomic properties of cfDNA. These findings demonstrate the potential of genome-wide epigenomic modelling as a clinically translatable paradigm for diagnostics and precision medicine, though prospective validation in larger, more diverse cohorts will be required to confirm clinical utility.

genomics↗

Time is encoded by methylation changes at clustered CpG sites

Age-dependent changes in DNA methylation allow chronological and biological age inference, but the underlying mechanisms remain unclear. Using ultra-deep sequencing of >300 blood samples from healthy individuals, we show that age-dependent DNA methylation changes are regional and occur at multiple adjacent CpG sites, either stochastically or in a coordinated block-like manner. Deep learning analysis of single-molecule patterns in two genomic loci achieved accurate age prediction with a median error of 1.46-1.7 years on held-out human blood samples, dramatically improving current epigenetic clocks. Factors such as gender, BMI, smoking and other measures of biological aging do not affect chronological age inference. Longitudinal 10-year samples revealed that early deviations from epigenetic age are maintained throughout life and subsequent changes faithfully record time. Lastly, the model inferred chronological age from as few as 50 DNA molecules, suggesting that age is encoded by individual cells. Overall, DNA methylation changes in clustered CpG sites illuminate the principles of time measurement by cells and tissues, and facilitate medical and forensic applications. O_FIG O_LINKSMALLFIG WIDTH=174 HEIGHT=200 SRC="FIGDIR/small/626674v1_ufig1.gif" ALT="Figure 1"> View larger version (46K): org.highwire.dtl.DTLVardef@c4dd52org.highwire.dtl.DTLVardef@9e6e62org.highwire.dtl.DTLVardef@160f8c3org.highwire.dtl.DTLVardef@16be7a3_HPS_FORMAT_FIGEXP M_FIG C_FIG

genomics↗

wgbstools: A computational suite for DNA methylation sequencing data representation, visualization, and analysis

Next-generation methylation-aware sequencing of DNA sheds light on the fundamental role of methylation in cellular function in health and disease. These data are commonly represented at a single CpG resolution, while single-molecule fragment-level analysis is often overlooked. Here, we present wgbstools, an extensive computational suite tailored for methylation sequencing data. wgbstools allows fast access and ultra-compact anonymized representation of high-throughput methylome data, obtained through various library preparation and sequencing methods. Additionally, wgbstools contains state-of-the-art algorithms for genomic segmentation, biomarker identification, genetic and epigenetic data integration, and more. wgbstools offers fragment-level analysis and informative visualizations, across multiple genomic regions and samples.

bioinformatics↗

Atlas of imprinted and allele-specific DNA methylation in the human body

Allele-specific DNA methylation, determined genetically or epigenetically, is involved in gene regulation and underlies multiple pathologies. Yet, our knowledge of this phenomenon is partial, and largely limited to blood lineages. Here, we present a comprehensive atlas of allele-specific DNA methylation, using deep whole-genome sequencing across 39 normal human cell types. We identified 325k genomic regions, covering 6% of the genome and containing 11% of all CpG sites, that show a bimodal distribution of methylated and unmethylated molecules. In 34K of these regions, genetic variations at individual alleles segregate with methylation patterns, thus validating allele-specific methylation. We also identified 460 regions showing parentally-imprinted methylation, the majority of which were not previously reported. Surprisingly, sequence-dependent and parent-dependent methylation patterns are often restricted to specific cell types, revealing unappreciated variation in the human allele-specific methylation across the human body. The atlas provides a resource for studying allele-specific methylation and regulatory mechanisms underlying imprinted expression in specific human cell types. HighlightsO_LIA comprehensive atlas of allele-specific methylation in primary human cell types C_LIO_LI325k genomic regions show a bimodal pattern of of hyper- and hypo-methylation of DNA C_LIO_LIAllele-specific methylation at 34k genomic regions C_LIO_LITissue-specific effects at known imprinting control regions (ICRs) C_LIO_LI100s of novel loci exhibiting parentally-imprinted methylation C_LIO_LIParentally-imprinting methylation is often cell-type-specific C_LI

genomics↗

Epigenetic liquid biopsies reveal elevated vascular endothelial cell turnover and erythropoiesis in asymptomatic COVID-19 patients

The full spectrum of tissues affected by SARS-CoV-2 infection is crucial for deciphering the heterogenous clinical course of COVID-19. Here, we analyzed DNA methylation and histone modification patterns in circulating chromatin to assess cell type-specific turnover in severe and asymptomatic COVID-19 patients, in relation to clinical outcome. Patients with severe COVID-19 had a massive elevation of circulating cell-free DNA (cfDNA) levels, which originated in lung epithelial cells, cardiomyocytes, vascular endothelial cells and erythroblasts, suggesting increased cell death or turnover in these tissues. The immune response to infection was reflected by elevated B cell and monocyte/macrophage cfDNA levels, and by evidence of an interferon response in cells prior to cfDNA release. Strikingly, monocyte/macrophage cfDNA levels (but not monocyte counts), as well as lung epithelium cfDNA and vascular endothelial cfDNA, predicted clinical deterioration and duration of hospitalization. Asymptomatic patients had elevated levels of immune-derived cfDNA but did not show evidence of pulmonary or cardiac damage. Surprisingly, these patients showed elevated levels of vascular endothelial cell and erythroblast cfDNA, suggesting that sub-clinical vascular and erythrocyte turnover are universal features of COVID-19, independent of disease severity. Epigenetic liquid biopsies provide non-invasive means of monitoring COVID-19 patients, and reveal sub-clinical vascular damage and red blood cell turnover.

microbiology↗

Cell-free, methylated DNA in blood samples reveals tissue-specific, cellular damage from radiation treatment

Radiation therapy is an effective cancer treatment although damage to healthy tissues is common. Here we characterize the methylomes of healthy human and mouse tissues to establish sequencing-based, cell-type specific reference DNA methylation atlases. Identified cell-type specific DNA blocks were mostly hypomethylated and located within genes intrinsic to cellular identity. Cell-free DNA fragments released from dying cells into the circulation were captured from serum samples by hybridization to CpG-rich DNA panels. The origins of the circulating DNA fragments were inferred from mapping to the established DNA methylation atlases. Thoracic radiation-induced tissue damages in a mouse model were reflected by dose-dependent increases in lung endothelial, cardiomyocyte and hepatocyte methylated DNA in serum. The analysis of serum samples from breast cancer patients undergoing radiation treatment revealed distinct tissue-specific epithelial and endothelial responses to radiation across multiple organs. Strikingly, patients treated for right-sided breast cancers also showed increased hepatocyte and liver endothelial DNA in the circulation indicating the impact on liver tissues. Thus, changes in cell-free methylated DNA can uncover cell-type specific effects of radiation and provide a quantitative measure of the biologically effective radiation dose received by healthy tissues. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=174 SRC="FIGDIR/small/487966v3_ufig1.gif" ALT="Figure 1"> View larger version (46K): org.highwire.dtl.DTLVardef@129f229org.highwire.dtl.DTLVardef@d999bcorg.highwire.dtl.DTLVardef@1fc3543org.highwire.dtl.DTLVardef@10cebc2_HPS_FORMAT_FIGEXP M_FIG C_FIG

cancer biology↗

A human DNA methylation atlas reveals principles of cell type-specific methylation and identifies thousands of cell type-specific regulatory elements

DNA methylation is a fundamental epigenetic mark that governs chromatin organization, cell identity, and gene expression. Here we describe a human methylome atlas, based on deep whole-genome bisulfite sequencing allowing fragment-level analysis across thousands of unique markers for 39 cell types sorted from 207 healthy tissue samples. Replicates of the same cell-type are >99.5% identical, demonstrating robustness of cell identity programs to genetic variation and environmental perturbation. Unsupervised clustering of the atlas recapitulates key elements of tissue ontogeny, and identifies methylation patterns retained since gastrulation. Loci uniquely unmethylated in an individual cell type often reside in transcriptional enhancers and contain DNA binding sites for tissue-specific transcriptional regulators. Uniquely hyper-methylated loci are rare and are enriched for CpG islands, polycomb targets, and CTCF binding sites, suggesting a novel role in shaping cell type-specific chromatin looping. The atlas provides an essential resource for interpretation of disease-associated genetic variants, and a wealth of potential tissue-specific biomarkers for use in liquid biopsies. Summary paragraphDNA methylation, a fundamental epigenetic mark, governs chromatin organization and gene expression1, thus defining the molecular identity of cells and providing a window into developmental processes with wide-ranging physiologic and clinical ramifications. Current DNA methylation datasets have limitations, typically including only a fraction of methylation sites, many from cell lines that underwent massive changes in culture or from tissues containing unspecified mixtures of cells2-6. We present a human methylome atlas based on deep whole-genome bisulfite sequencing of 39 sorted, primary cell types and use this dataset to address fundamental questions in developmental biology, physiology and pathology. Biological replicates are >99.5% identical, demonstrating unappreciated robustness to genetic variation and environmental perturbations. Clustering recapitulates key elements of tissue ontogeny, identifying methylation patterns retained since gastrulation. Loci uniquely unmethylated in individual cell types identify novel transcriptional enhancers and are enriched for tissue-specific transcription factors binding motifs. In contrast, loci uniquely hyper-methylated in specific cell types are rare, enriched for CpG islands and polycomb targets, and overlap CTCF binding sites, suggesting a novel role in shaping cell-type-specific chromatin looping. Finally, the atlas facilitates fragment-level deconvolution of tissue and plasma methylomes across thousands of cell-type specific regions to quantify their individual components at unprecedented resolution. The human cell-type-specific methylation atlas provides an essential resource for studying gene regulation by defining cell-type-specific distal enhancers and regulators of 3D organization, for identifying pathological changes in DNA methylation, and for the interpretation of methylation-based liquid biopsies. O_FIG_DISPLAY_L [Figure 1] M_FIG_DISPLAY C_FIG_DISPLAY O_LIA deep methylation atlas of 39 human cell types, sorted from healthy samples C_LIO_LIMethylomes record developmental history of cells C_LIO_LIThousands of novel cell type-specific methylation markers C_LIO_LIHypo-methylation uncovers cell type-specific regulatory map of distal enhancers C_LIO_LIHyper-methylation across CTCF sites C_LIO_LICell type-specific biomarkers facilitate fragment-level deconvolution of tissues and cfDNA C_LI

genomics↗

Remote immune processes revealed by immune-derived circulating cell-free DNA

Blood cell counts often fail to report on immune processes occurring in remote tissues. Here we use immune cell type-specific methylation patterns in circulating cell-free DNA (cfDNA) for studying human immune cell dynamics. We characterized cfDNA released from specific immune cell types in healthy individuals (N=242), cross sectionally and longitudinally. Immune cfDNA levels had no individual steady state as opposed to blood cell counts, suggesting that cfDNA concentration reflects adjustment of cell survival to maintain homeostatic cell numbers. We also observed selective elevation of immune-derived cfDNA upon perturbations of immune homeostasis. Following influenza vaccination (N=92), B-cell-derived cfDNA levels increased prior to elevated B-cell counts and predicted efficacy of antibody production. Patients with Eosinophilic Esophagitis (N=21) and B-cell lymphoma (N=27) showed selective elevation of eosinophil and B-cell cfDNA respectively, which were undetectable by cell counts in blood. Immune-derived cfDNA provides a novel biomarker for monitoring immune responses to physiological and pathological processes that are not accessible using conventional methods.

genomics↗

Comparative Parallel Multi-Omics Analysis During the Induction of Pluripotent and Trophectoderm States

Following fertilization, totipotent cells divide to generate two compartments in the early embryo: the inner cell mass (ICM) and trophectoderm (TE). It is only at the 32-64 -cell stage when a clear segregation between the two cell-types is observed, suggesting a T-shaped model of specification. Here, we examine whether the acquisition of these two states in vitro by nuclear reprogramming share similar dynamics/trajectories. We conducted a comparative parallel multi-omics analysis on cells undergoing reprogramming to Induced pluripotent stem cells (iPSCs) and induced trophoblast stem cells (TSCs), and examined their transcriptome, methylome, chromatin accessibility and activity and genomic stability. Our analysis revealed that cells undergoing reprogramming to pluripotency and TSC state exhibit specific trajectories from the onset of the process, suggesting V-shaped model. Using these analyses, not only we could describe in detail the various trajectories toward the two states, we also identified previously unknown stage-specific reprogramming markers as well as markers for faithful reprogramming and reprogramming blockers. Finally, we show that while the acquisition of the TSC state involves the silencing of embryonic programs by DNA methylation, during the acquisition of pluripotency these specific regions are initially open but then retain inactive by the elimination of the histone mark, H3K27ac.

developmental biology↗

Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease

Methylation patterns of circulating cell-free DNA (cfDNA) contain rich information about recent cell death events in the body. Here, we present an approach for unbiased determination of the tissue origins of cfDNA, using a reference methylation atlas of 25 human tissues and cell types. The method is validated using in silico simulations as well as in vitro mixes of DNA from different tissue sources at known proportions. We show that plasma cfDNA of healthy donors originates from white blood cells (55%), erythrocyte progenitors (30%), vascular endothelial cells (10%) and hepatocytes (1%). Deconvolution of cfDNA from patients reveals tissue contributions that agree with clinical findings in sepsis, islet transplantation, cancer of the colon, lung, breast and prostate, and cancer of unknown primary. We propose a procedure which can be easily adapted to study the cellular contributors to cfDNA in many settings, opening a broad window into healthy and pathologic human tissue dynamics.

molecular biology↗